If you’ve used ChatGPT before, it can be seen that it has a certain writing style. It tends to use words such as “cornerstone” a lot, and certain types of sentence structures. However, it’s definitely possible for humans to write in the way that ChatGPT does as well. I find that as I read AI-created content more, I adopt the AI style of writing into my own work, which could make it appear to be AI-generated.
Some companies claim they can actually detect AI. Usually, this may involve using some AI themselves. So if they don’t like AI, why are they using AI to detect it?
Human-written content is often said to be AI-generated. For instance, a recent study found that some of the most popular AI detection tools have false positive rates. For example, plagiarism detective service Turnitin detected articles as fully AI-generated, even when 20% or less was written by AI. Many students have already reported problems with teachers believing their essays are AI-written due to false analysis.
The Commercial Agenda
The commercial interests behind AI detection tools cannot be ignored. Companies are quick to monetize these…